Sleep and Ventricular Premature -...

Sleep and Ventricular Premature Beats

By BERNARD LowN, M.D., MARK TYKOCINSKI, ANDRE GARFEIN, AND PHILLIP BROOKS

SUMMARYSixty-nine 24-hour electrocardiographic monitoring sessions were conducted in 54 ambulatory

subjects. Thirty-one had coronary heart disease (CHD); 11 had miscellaneous heart ailments;and 12 were free of any heart disease. Monitoring was accomplished with a miniaturized cassetterecorder and ventricular premature beats (VPBs) were analyzed as to incidence and grade. Theoccurrence of VPBs during sleep hours was compared to the awake state.

In 22 patients, the incidence of VPBs was reduced by at least 50% during sleep. An additional13 patients showed a reduction of 25 to 50%. If patients free of ectopic activity during 24-hourmonitoring sessions are excluded from analysis, then in 35 of 45 patients, or in 78%, sleep wasassociated with a lowered occurrence of ventricular extrasystoles. During sleep the VPB grade waslikewise lowered. Thus the mean grade for 45 patients changed from 2.75 while awake to 1.78while asleep. It is of interest that in a number of these patients, trials of various antiarrhythmicdrugs were less effective than sleep in reducing the incidence and grade of VPBs. It is concludedthat treatment of sporadically occurring ventricular ectopic activity in some patients may requireattention to the neurophysiologic trigger rather than the cardiac target.

Additional Indexing Words:Monitoring Ventricular ectopic activity Ventricular tachycardia Neural activity

VENTRICULAR ectopic beats are sporadic andseemingly random in occurrence. The reason

for their emergence precisely at any moment intime remains obscure even in patients withsignificant heart disease; in most instances, noimmediate provocative factor can be identified. Thedevelopment of extrasystoles without apparentcause in: healthy subjects suggests that there may bemediating agencies extrinsic to the heart. Physicianshave long been aware that psychologic and neuralfactors may affect heart rhythm. Irregularities inpulse due to emotion were already appreciated byCalen in the Roman era. The passage of twomillenia has provided much anecdotal informationrelating cardiac arrhythmia to psychologic factors,but hard clinical facts continue to be meager.

From the Cardiovascular Research Laboratories, Depart-ment of Nutrition, Harvard School of Public Health; theLevine Cardiac Unit, Cardiovascular Division, Departmentof Medicine, Peter Bent Brigham Hospital, Boston, and theResearch Laboratories, American Optical Company, Fram-ingham, Massachusetts.

Supported in part by grant MH-21384 from the NationalInstitute of Mental Health and grant HL-14602 from theNational Heart and Lung Institute of the National Institutesof Health, U. S. Public Health Service.

Address for reprints: Bernard Lown, M.D., Department ofNutrition, Harvard School of Public Health, 665 HuntingtonAvenue, Boston, Massachusetts 02115.

Received April 16, 1973; revision accepted for publicationJune 7, 1973.

Circulation, VTolum6 XLVIII, October 1973

The sleep-wakefulness cycle represents a markeddiurnal variation in behavioral and neural activity.Yet day-night alterations in cardiac rhythm havenot been adequately studied. This is surprising, for50 years ago the noted physiologist MacWilliamcalled attention to the fact that sleep was at timesassociated with profound cardiac alterations.' Heemphasized that sleep represents a continuous shiftbetween disturbed and sound phases and that"circulatory changes in disturbed sleep are some-times so very pronounced that it is remarkable theyshould so long have escaped observation."The present investigation addresses itself to one

preliminary aspect, namely, the diurnal variation inthe frequency and pattern of ventricular prematurebeats (VPBs) in patients with and without heartdisease.

Material and MethodsThe Population

Sixty-nine 24-hr electrocardiographic monitoringsessions were conducted on 54 ambulatory subjects.Heart disease was present among 42 and is detailed intable 1. Of the 31 patients with coronary heart disease(CHD), 28 had sustained a previous documentedmyocardial infarction while three were afflicted withangina pectoris. Eleven patients, six males and fivefemales, had diverse cardiac disease, while 12 subjectswere free of organic cardiovascular problems.The mean age for those with CHD was 55, with a

range of 33 to 76 years. Eleven of these patients hadsustained myocardial infarction in the preceding 12

691

by guest on June 20, 2018http://circ.ahajournals.org/

Dow

nloaded from

http://circ.ahajournals.org/

LOWN ET AL.

Table 1Types of Heart Disease Among 54 Monitored Patients

Number Male Female

Coronary heart disease 31 28 3

Miscellaneous heart disease 11 6 5Rheumatic 4Cardiomyopathy 2Ruptured chordae 1Atrial septal defect 1Hypertensive 1Undetermined 2

No heart disease 12 8 4

months, five within two years, four within three yearsand in the remaining eight the interval exceeded threeyears. In 21 patients the monitoring was routine andnot occasioned by the presence of known or troublingarrhythmia. In 10 patients, monitoring was institutedbecause of arrhythmia; in three of the 10 patients themechanism was atrial fibrillation; in three VPBs werepresent; in two there was paroxysmal ventriculartachycardia; one had atrial flutter and one hadexperienced a single episode of ventricular fibrillation.Twelve of the patients were receiving antiarrhythmicdrugs: three, procaine amide; six, quinidine sulfate; andthree, propranolol. The latter drug was being admini-stered for severe angina pectoris. None of these drugswas withheld during the study.The 11 patients with heart disease other than CHD

ranged in age from 33 to 68 with a mean of 54.2 years.Eight were monitored routinely, while three patientswere monitored because of one of the followingrecurring disorders: frequent VPBs, ventricular tachy-cardia, or ventricular fibrillation.The 12 patients without heart disease, ranging from

26 to 68 years, had a mean age of 48. All weremonitored because of some symptoms such as palpita-tion, dizziness, weak or fainting spells, possiblyascribable to arrhythmias or the known presence ofVPBs.

Monitoring EquipmentContinuous electrocardiographic recordings were

obtained by means of a portable cassette tape unit.*The recorder measures 13.5 x 10.3 x 4.8 cm andweighs with batteries and carrying pouch 680 g (1.5lbs). The unit is powered by two 7-volt type TR 135mercury batteries. It is driven by an AC synchronousmotor at a tape speed of approximately 2 mm/sec. FMrecording is utilized to obtain a frequency response of0.3 to 25 Hz. Standard Phillips type C-120 cassettespermit data acquisition for a period of 24 hr on eachside of the cassette.

Patients were coupled to the monitoring apparatuswith standard disposable silver-silver chloride elec-trodes, employing a modified lead II placement. One

lead was positioned in the second right intercostalspace at the sternal border and the other lead was inthe fifth left intercostal space at the left anterior axillaryline. The compact recorder was attached to a belt wornaround the waist and did not interfere with strenuousactivity or even vigorous exercise. An hourly event cardwas provided for checking off activities such as eating,smoking, exercise, bowel movements, the occurrenceand type of symptoms and drug ingestion. Theimportance of recording the time of going to bed, theapproximate time of falling asleep, and the precise timeof waking up was emphasized to each participant in thestudy. The patient was advised to omit sleepingmedication during the monitoring session; all but fivedid so. No other drugs were prescribed. Patients were

instructed to proceed with their routine activities andnot to avoid their usual daily stresses.

Data Reduction

Analysis of the 24-hr continuous ECG records was

accomplished with a high speed playback analyzerrunning at 60 times real time, and a low speed paperrecorder. Tapes were played back on an Auricordcassette deck and analyzed by a trained operator.Audio-visual techniques using an R wave triggeredoscilloscopic display and an ECG coupled sound systemenabled detection of VPBs. The technician employed aspecial hand counter to record cumulatively thefrequency of VPBs per minute on a trend record whichsimultaneously charted heart rate. A slow speed displayand printout permitted clarification of questionablehigh speed observations. When an arrhythmia wasdetected that portion of the tape was rerun at normalspeed and recorded. The printed record of all thearrhythmias was checked by a physician. Cassetteswere not erased and were replayed by two othertechnicians for verification of the initial data acquisi-tion.

Frequent VPBs exceeding one per minute cannot beaccurately analyzed with the high speed system. Thisnecessitated employment of additional methods for datareduction. Various sampling techniques were devel-oped. The cassettes were replayed twice real time,permitting 24 hr tape review in 12 hr. Tandemrecycling clocks were preset to activate the paper chartprintout at fixed intervals for fixed durations. Thesewere run during the night, thus freeing the playbackanalyzer for daytime use.

All tapes were scanned at 60 times real time andthen classified into one of three categories:

1) Low VPB frequency-VPBs occurring at lessthan one per minute during each hour of the monitoringsession. In this category the initial scanning provedadequate for precise definition of frequency and gradeof ectopic activity. These tapes were not processedfurther.

2) High VPB frequency-VPBs occurring with anincidence in excess of three per minute. These tapeswere reviewed with the sampling technique. Thesampling intervals consisted of 30 sec printout every 20min or 15 sec printout every 10 min. This provided 36min real time printout for a 24 hr period of monitoringor 2.5% of the available data. Both of these methods

Circulation, Volume XLVIIIJ, October 1973

* Developed by American Optical Company, Framingham,Massachusetts.

692


Dow

nloaded from


SLEEP AND VPBs

Table 2

Incidence of Ventricular Premature Beats (VPBs) During Sleeping Compared to Waking Hoursin 54 Patients, Monitoring Sessions

Alteration in incidence of VPBs

Substantial Marginaldecrease decrease No change Increase No VPBs Totals

Patients Patients/Patients Sessions Sessions* Patients Sessions Sessions* Patients Sessions Patients Sessions

Coronary heartdisease

Miscellaneousheart disease

No heart diseaseTotal

13 17 10 4 4 1 7 8 35t 40

54

22

57

29

1

213

228

42

10

1

0

2

2514

2515

1113159

131669

* When there is identity between numbers of patients and sessions, these are not separated.t In four patients monitored two sessions each there was a difference in sleep-wake findings. Each result is

presented accounting for the number exceeding the 31 CHD patients studied.1 One patient monitored twice showed a discrepancy of result and thus appears twice.

yielded identical results as to grade and frequency ofectopics. On the basis of these samples the VPBincidence per hour was estimated.

3) Moderate VPB frequency-VPBs ranging inincidence from one to three per minute and highlyrandom in occurrence. This required more tediousprocessing. In such instances, scanning techniquesestablished the time of development of VPB clustersand sampling time was then lengthened to record inreal time their occurrence.

Data AnalysisFor each monitoring hour the incidence and grade of

VPBs were determined. The grading system has beenpreviously described2 and assigns VPBs into one of thefollowing categories: 1) occasional VPBs, 2) frequentVPBs (greater than 30/hr or 1/min), 3) multiformVPBs, 4) repetitive VPBs, a) pairs (2 consecutiveectopics), b) runs (3 or more in sequence), 5) earlyVPBs (R on T).

Sleep periods were determined from the patient'sactivity card. The recorded trend in heart ratereduction provided supporting evidence that sleepoccurred. Since there was no certainty as to the precisetime of falling asleep, the hour in which the patientindicated that this occurred was excluded from thesleep-awake analysis.The average hourly incidence during waking hours as

well as the highest VPB grade achieved was deter-mined. A similar determination was carried out for thehours the patient slept. Sleep-awake periods were

compared if there were at least 10 VPBs during one ofthese diurnal cycles. For a change in VPB frequency inthe two periods to be regarded substantial, the average

hourly incidence during waking hours had to differ by50% or more from the average obtained during sleephours. Changes from 25 to 50% were regarded as

marginal. For a change in grade to be deemedsignificant, alteration by at least one grade was

required, with 4a and 4b considered to be differentgrades. For statistical calculation 4b was assigned a

numerical value of 4.5. Student's paired t-test was usedfor statistical analysis.

ResultsGeneral Findings

Of the 54 patients, 45 (83.3%) demonstrated some

ventricular ectopic activity during at least one 24-hrmonitoring session. VPBs were observed in 54 ofthe 69 sessions (78.3%). Sleep was associated with a

reduction of 50% or greater in the incidence of VPBsamong 22 patients during 29 monitoring sessions. Adecrease between 25 and 50% was noted in 13additional patients. Thus 35 of the 54 patients

'able 3

Change in Grade of VPBs During Sleep in 45 Patients with Ectopic Activity During 24-HourMonitoring

Type of Patients Sessions Mean Gradeheart disease (number) (number) Awake Asleep P value

Coronary heartdisease 28 32 2.6 1.6 < 0.001

Miscellaneous heartdisease 9 11 3.5 2.8 < 0.1

No heart disease 8 11 2.4 1.2 < 0.025

Circulation, Volume XLVIII, October 1973

Diagnosis

693


Dow

nloaded from


LOWN ET AL.

Table 4Grade of VPBs Observed as well as Maximum GradeReached During Each of 40, 24-Hour Monitoring Ses-sions, in 31 Ambulatory Patients with CHD

Grade of VPB

01234a4b5

Number of sessionsGrade Maximum grade

observed reached

8171315724

61209724

(64.8%) exhibited a lessening VPB frequencyduring sleep (table 2). These findings become morestriking if the nine patients who were free ofarrhythmia throughout the 24-hr monitoring sessionare excluded from the analysis. Of the 45 patientswho exhibited ectopic activity, 35 or 77.7% showed adecrease. In nearly half of these patients the meanhourly incidence during sleep was reduced by 50%or more as compared to those patients observedduring waking hours. The tendency for sleep to beassociated with a reduction in VPBs was observedin patients with far advanced heart disease andcongestive failure as well as in subjects without anycardiac ailments.

In addition to the reduction in VPB incidenceduring sleep there was also a lessening in grade(table 3). Thus in 54 monitoring sessions among 45patients with ectopic activity, there was a change inmaximum grade from 2.74 while awake to 1.76during sleep (P < 0.001). The extent of gradealterations in the three groups of patients is detailed

in table 3. In 12 of the 54 sessions (22%) VPBsdisappeared entirely during sleep.Coronary Heart Disease (CHD) PatientsThe 31 patients with CHD were monitored 40

sessions for a total of 960 hours. The distribution ofVPBs by grade as well as by maximum gradeachieved during a single session is presented intable 4. Of interest is that in none was grade 2 thehighest grade achieved. This is accounted for by thefact that whenever VPBs were frequent, they wereinvariably multiform or repetitive.Among the 28 CHD patients with ventricular

ectopic activity, 13 patients during 17 monitoringsessions showed a decreased incidence of 50% orgreater during sleep (table 5). There was alsoreduction in highest grade observed during wakinghours, 2.6, to 1.6 during sleep (P < 0.001). Thefollowing patient is representative of this group:A 63-year-old man, C.E., had sustained an acute

transmural myocardial infarction two years earlier.While awake he experienced frequent and multi-form VPBs. He was not receiving any antiarrhyth-mic drugs. During 15 waking hours ectopic activitywas never less than 2/min and was frequentlybigeminal in pattern. During nine of these hours thefrequency of VPBs exceeded 15/min. However,during five of his sleeping hours, ectopic activitywas entirely absent (fig. 1). When he was

25

20k

15sTable 5Change in Maximum Grade of VPBs During Sleep asCompared to Awake Hours. Analysis Based on 22 Pa-tients (29 Monitoring Sessions) Who Showed a 50%or Greater Reduction in Incidence of Ectopic Activity.

oh

5Number of sessions with

reduction during sleep by:Monitoring 3 2 1 0

Diagnosis sessions grades grades grade grade

Coronary heartdisease (13)*

Miscellaneousheart disease(5)

No heart disease(4)

Total* In ( ) are number

0O

17 3 3 5 6

}fl.ifl310 12 2 4 6 8 10 12 2 4 6 8

-AM- PM " AM -A

AWAKE

Figure 15 1 2 1 1ASLEEP

7 2 1 2 2 (C.E.-Patient with CHD.) During five of eight sleep hours,29 7 6 8 8 the incidence of VPBs was consistently less than 1/min;

during all waking hours, the arrhythmia was of Grade 2p of patients in each group. with frequent bigeminy.


__ _. b.s. @. w { ,--.--.k.::k:;:k:,

0..

694


Dow

nloaded from


SLEEP AND VPBs

remonitored three weeks later, an identical diurnalpattern in VPB distribution was observed (fig. 2).Peak ectopic activity occurred during early morninghours, reaching a level of 25/min. Thereafter VPBsdecreased and during five sleep hours no ectopicactivity was encountered.

In all patients progressive reduction in heart ratewas observed during sleep; an example is illustratedin figure 3. The frequency of VPBs did not appearto correlate with the minute-to-minute variation inheart rate. In a number of these patients when theheart rate reached its lowest level toward late sleep,there was a resumption or increase in incidence ofVPBs. Ectopic morphology frequently altered dur-ing sleep. This was observed in 10 of the 17 sessionsdemonstrating a greater than 50% reduction in VPBactivity during sleeping as compared to wakinghours (fig. 4).

In 10 patients with CHD subjected to 10 mon-itoring sessions, the VBP incidence nocturnallywas reduced by less than 50% compared to the meanfor waking hours. None of these patients experi-enced advanced grades of arrhythmia. Ventricularectopic activity was at a low level, rarely exceeding2/hr and occurred only in a fraction of the wakinghours. VPBs were noted in 29% of the 16 wakinghours, compared to 12% of seven sleep hours.

In four patients no change in incidence or gradeof VPBs was noted during sleep. These patients hada high frequency of ectopics: in one, grade 3; intwo, grade 4a; and in one, grade 4b. Only one pa-tient showed an increase of ectopic activity during

sleep. The reason for this is not clear. All five ofthese patients had taken sleeping pills and were theonly CHD patients to do so.

Miscellaneous Heart Disease

Thirteen monitoring sessions were conductedamong this diverse group of 11 patients (tables 1and 2). Five of these patients were receivingdigitalis and diuretic drugs. Six of nine patientswith arrhythmia showed a reduction in incidence ofVPBs during sleep. The mean grade was alteredfrom 3.5 while awake to 2.5 while asleep(P 0.05). In three patients the reduction was byat least two grades (table 5). One of the twopatients afflicted with cardiomyopathy is of particu-lar interest.

R.B. was a 62-year-old man with a long history ofheavy alcohol consumption. Congestive heart fail-ure developed four years earlier and progressed torecurring pulmonary edema. Cardiac catheteriza-tion showed left ventricular (LV) end-diastolicpressure of 20 mm Hg, a cardiac index of 1.5L/min, an ejection fraction of 20%, and uniformlypoor ventricular wall motion. Coronary angiogra-phy revealed none of the major vessels to benarrowed by more than 50%. Frequent but asymp-tomatic ventricular ectopic activity was consistentlyobserved during daytime monitoring, with a patternof ventricular bigeminy: the highest grade of VPBsreached was 4a. He had experienced difficulty insleeping, falling off to sleep only after 2:00 a.m.During the ensuing four hours of sleep, though the

Time

Y~~~~~~~~~~~~~~~~~~~~~~L

11:00 AM

8:00 PM +2< . , . _ . . . _ _ _ + . .~~~~~~~~~~~~~~~-1l--4=;-- I-0t---t--t=-- F t { s~~~~~~~~~~~~~~~~~~~~LLt_ tl .--t---- l - 1--t- - - -t- - - -[ -t ----- { i - - t i

12:30~~~~~~~~~~~~~~ AM12.5 mm/Sec

Figure 2

(C.E.-same patient as shown in fig. 1.) In the morning during second monitoring session both ventricularand atrial ectopic beats prevailed. In early evening (8:00 p.m.) the VBP's were multiform, bigeminal andquadrigeminal. During sleep extrasystoles disappeared. The sinus rate at these three times was 74, 78, and78/min, respectively.


695


Dow

nloaded from


LOWN ET AL.

Awake Asleep Awake[-~~~~~~~~ --11501

125HE&ARr

rate/lm/n 10075

50

160H

VPB/Hr120H

80

40

LI] Multiform

Couplets

k:JH H12 2 4 6 8 10 12 2 4 6 8 10 12

PM I AM -Figure 3

(J.B.-43-year-old man with old diaphragnatic transmural infarction.) With onset of sleep Grade 2 VPBswere abolished. There was a progressive slowing in heart rate during the night. Upon awaking there was arecurrence of VPBs.

heart rate continued to be rapid, VPB frequencywas reduced to 35/hr and did not exceed grade 1.At 6:00 a.m., when the lowest heart rate wasreached, VPBs disappeared entirely only to increaserapidly upon awaking shortly thereafter (fig. 5).

Subjects Without Heart DiseaseThis group included 12 subjects, all of whom

were monitored because of the presence of VPBs orof symptoms suggestive of arrhythmia. Five wereentirely free of ectopic activity during the 24-hrmonitoring period. Four of the remaining sevenpatients showed a substantial decrease in VPBsduring sleep. In three patients VPBs were complete-ly absent during sleep. Of interest is the fact that inthe absence of heart disease, unlike in patients withCHD, frequent VPBs were not associated withhigher grades of arrhythmia. In five of 11monitoring sessions, grade 2 was not accompaniedby multiformity or repetitive firing. One patientwho was monitored three times at weekly intervalsdeserves brief comment.

SB., a 34-year-old asymptomatic mother of three,was observed to have transient T wave inversions

and frequent ectopic beats following a cholecystec-tomy one year earlier. There was no evidence ofheart disease. The resting electrocardiogram, exceptfor frequent VPBs, was normal. Maximal treadmillexercise testing was negative. She was initiated on4.0 g of procaine amide daily resulting in a bloodlevel of 6.0 ,ugm/ml without effect on ectopicactivity. An increase in procaine amide to 6.0 gdaily provoked adverse reactions. Quinidine sulfate0.4 g four times daily resulted in a blood level of 4.5ggm/ml and reduced the incidence of VPBs. Theaddition of propranolol, initially in a dose of 80 mgand then 160 mg daily (blood level 68 ng/ml),appeared to attenuate the arrhythmia further (fig.6) but did not abolish ectopic activity. However,during each of the three monitoring sessions, VPBswere markedly reduced during sleep and wereentirely absent two nights.

Reproducibility of Sleep-Wake ChangesTo determine reproducibility of sleep-wake

changes, 13 subjects were monitored a total of 28sessions. The mean interval between monitoringsessions was one week. The group consisted of nine


I1 1;

696

I1 1 1 Il l A 1


Dow

nloaded from


SLEEP AND VPBs

Time Rate .IJI 7

1:00 PM

75 --

6:00 PM L

145 -

_i _ _

r1_7. -

a

a*

......: -

11 tL 1AI-I1

ihEA

900 PM Xz1iA

76

300AM

50so~ L~?OOAM

I1

75

I.'' 1, 1 1---

-1 1 - L i -

t . 1 I

..1.. .. .. . . .1

, 1 !-- 1:- -} , '1- 3i-Xi _ _L

Me:F z, _ lg '-:i' L0.t-t: |

he

mm

.I*..ImFigure 4

(J.B.-same patient as shown in fig. 3.) Frequent and multiform VPBs disappeared during vigorous game oftennis at 6:00 p.m. with heart rate acceleration to 145/mim. Couplets were observed while he was restingand reading qutietly at home. During sleep the heart rate decreased to 50/min, a new ectopic morphologydeveloped-a junctional focus with aberration cannot be excluded (paper speed 12.5 mm/min).

females and four males. CHD was present in nine,one had cardiomyopathy, and three were withoutheart disease. One patient was free of VPBs duringtwo 24-hr monitoring sessions. In eight of the 13patients, or in 61.5%, the same maximum VPB gradewas observed upon repeated 24-hr monitoring. Aconsistent decrease in VPB incidence during sleepin successive monitoring occurred in six patients(46%). No reproducibility in sleep-wake comparisonwas observed in five patients (38.5%). One of these


patients with grade 4a arrhythmia on two monitor-ing sessions while awake showed a decrease duringthe first night, but on the second occasion, whensleep was interrupted by gas pains, the incidence ofVPBs remained unaltered.

Other ObservationsWhile a decrease in ventricular ectopic mechan-

isms was evident during sleep, atrial arrhythmiaappeared to increase. Atrial premature beats were

- - -,qv 7w

-11 1 - #*

M! t

a

697

19

.11L- 1 L- L_

-11,

T it1


Dow

nloaded from


LOWN ET AL.

25

20H

15

10

10 12 2 4 6 8 10 12 2 4 6 8FAM PM AM

-AWAKE AWAKEASLEEP

Figure 5

5

0'

95

90

85

80

75

70

Lu1

Kl

(R.B.-patient with cardiomyopathy.) Exhibited multiformVPBs during the majority of hours while awake. When thepatient finally slept, there was a striking reduction in VPBincidence. Though the heart rate decreased during sleep, itwas more rapid than during the majority of waking hours.

more numerous during sleep than during wakinghours. Many patients exhibited brief paroxysms ofsupraventricular tachycardia, even when sleepsubdued the frequency of ventricular ectopicactivity. Episodes of sinus bradyeardia and sinuspauses with shift to a junctional mechanismpunctuated by bursts of complex atrial ectopicrhythms were common. The precise definition ofthese nocturnal changes awaits further technologicdevelopment in methods for reducing monitoreddata relating to atrial and junctional arrhythmia.

DiscussionUntil a few decades ago sleep had been viewed

as a unique passive resting phase of the brain, "thediastole of the cerebral beat."3 Modern sleeppsychophysiology had its inception with the pio-neering investigations of Asserinsky and Kleitman4and Dement and Kleitman5 who discovered regular-ly recurring sleep cycles. These were distinguishedby rapid eye movements (REM) and EEG patternsassociated with alert wakefulness which correlatedwith the recall of dreams. The generalizationemerging from these findings is that during sleepthe brain, and the peripheral nervous system underits control, undergo continuous oscillations inactivity with a periodicity approximating 90 min-utes.

Sleep has been arbitrarily divided into five stagesdepending upon distinctive polygraphically re-corded physiologic activity. While a good deal ofdata has been accumulated relating stages of sleepto heart rate and blood pressure changes, there isvery limited information on the effects of sleep inpatients with cardiac disease. Nowlin et al.6correlated the occurrence of nocturnal angina withREM periods. During desynchronized REM sleepthey observed significant ST-segment depressionseven when the patient was not awakened byanginal discomfort. Karacan, Williams and Taylor7found no such association between nocturnalangina and the REM state; they did note thatpatients with angina experienced fewer shifts in thestages of sleep than control subjects.The literature on the effects of sleep on heart

rhythm is noteworthy for its poverty of meaningfulobservations. In one limited study of 13 subjectswith premature beats, ectopic activity appeared tobe more frequent during REM periods and sleepstages 3 and 4. Arrhythmias also occurred whileawaking from any stage.8 The prominence ofarrhythmia during deep sleep stages was ascribedto bradycardia. In a second study, 17 patients withVPBs were monitored electroencephalographicallyas well as for heart rhythm for a period of 6.5 hrwhile in a coronary care unit.9 Eight patients hadacute myocardial infarction, eight had diverse heartdisease and one had digitalis intoxication. Duringthe nocturnal monitoring, occurrence of ventricularand atrial extrasystoles did not vary in the differentstages of sleep or in the awake as compared toasleep periods. The highest frequency of VPBs wasnoted during sleep-wake transitions. No consistentrelation could be established between emergence ofectopics and altered heart rate. One additionalnotation has been published on sleep and cardiacarrhythmia involving a 48-year-old woman withpsychiatric problems, factitial skin ulcers, VPBs andrecurrent paroxysms of ventricular tachyeardia.10Various antiarrhythmic drugs were ineffective. Itwas observed repeatedly that when she slept allectopic activity disappeared.The focus of the present study was limited to

examining the incidence and grade of VPBs duringthe sleep period compared to the awake state. Areduction was observed in nearly 80% of 45 patientswho exhibited arrhythmia during daytime monitor-ing. Since onset of sleep could not be precisely andobjectively defined, this conclusion is in need ofcritical examination. The exclusion from the analy-sis of the transitional hour between wake and sleep


698


Dow

nloaded from


SLEEP AND VPBs

Asleep Awake

Asleep Awake B*- wl

- go!Y *

CAwa ke

.-

Asleep Awa ke* *

.. . .

. ..

.. ... . .... . .. ...

... .... .... .... .. I2 4 6 8 10 12 2 4 6 8 10 12 2 4

1--PM "- AM- " PM-iFigure 6

S.B. was monitored three successive 24-hr sessions at weekly intervals. With sleep there was nearly a totaldisappearance of VPBs. This change was greater than that observed following ingestion of antiarrhythmicdrugs, the time for which is indicated by asterisks. (A) Patient received 0.4 g of quinidine. (B) Quinidinewas continued, but 20 mg propranolol was added. (C) The dose of propranolol was increased to 40 mg.

stages enchanced accuracy of the comparison. It isunlikely that patients would have erred by more

than one hour in defining onset of sleep. If such an

error were made by each patient, which is highlyimprobable, and sleep had been adjudged to haveoccurred earlier than it actually did, the average

incidence of ectopic activity during sleep wouldhave been overestimated by about 14%. If, converse-

ly, sleep actually occurred later, the average

incidence of VPBs while awake would have beenunderestimated by about 7%. In either case, theerror would have tended to bias against rather thanfor the reported result.

Patients slept at home and in their own beds.This afforded a great advantage over the unnaturalCirculation, Volume XLVIII, October 1973

environment of a sleep laboratory. On the otherhand, it accounts for a major limitation of thepresent study, namely, the absence of objective dataon sleep and its stages. It is uncertain whether sleepwas consistently maintained during the periodindicated by the patient. However, if some of thepatients had slept poorly or not at all, there mighthave been an increase in incidence and grade ofVPBs during such periods. This would dilute theobservation that sleep reduced ventricular ectopicactivity.Of the 54 subjects studied, 42 had heart disease;

in the majority the process was stable and ofprolonged duration. The finding of a reduction inVPBs with sleep may not apply when the cardiac

Awake A* *

Awake

* *

20

10

0

201

1o0

OL

201

U L_

b... .4 go!

90m 10 ..

699

*

10o_

^ 1fl---.9

Z..


Dow

nloaded from


LOWN ET AL.

disease is acute and unstable. Indeed, a somewhatdifferent result was obtained among patientsconvalescing from acute myocardial infarction(Kerzner and Lown, unpublished observation). Ineight of 48 such patients (17%), there occurredeither aggravation of pre-existing arrhythmia ordevelopment of ectopic mechanism during sleep notrecorded while awake. In five of these eight, VPBsincreased either in frequency or grade during sleep.The remaining three patients showed ventriculararrhythmia only while asleep; one patient hadventricular tachycardia, a second had bigeminy,while the third had occasional VPBs.With sleep the heart rate decreases progressively

throughout the night reaching its nadir at about thesixth hour.'1 One might therefore have anticipated anocturnal increase rather than diminution in VPBs.Han et al.12 have demonstrated that ventricularectopic beats were more likely to develop in theexperimental animal when the basic rate was slow.This was true whether the precipitating agency ofectopic activity was coronary occlusion, suddenelevation in left ventricular pressure or hypother-mia. These investigators have also noted thattemporal dispersion of refractory period durationwas augmented by slow rates.'3 Thus bradyeardiawould also favor repetitive ectopic activity on thebasis of reentrant excitation. An association be-tween slow heart rates and ventricular arrhythmia isknown to occur in man under diverse clinicalcircumstances. Carotid sinus massage may provokeventricular extrasystoles in patients with CHD.14Digitalis intoxication-induced ventricular ectopicrhythms can be exposed by a diversity of measureswhich slow the basic frequency including vagalstimulation,'5' 16 administration of rauwolfia alka-loids,'7 beta adrenergic blocking drugs,'8 and theinjection of succinylcholine.19 In patients with atrialfibrillation, early ectopic beats develop morefrequently after relatively long ventricular cycles.20Complete heart block is associated with a highincidence of ventricular arrhythmia promptly sub-dued by accelerating the heart rate. In sleep, thisprevalent tendency associating slow heart rates withventricular ectopic activity appears to be dimin-ished or annulled.The reduction in VPBs during sleep was most

marked in patients without heart disease. Thusthree out of seven of these subjects lost all ectopicactivity while asleep. This was observed in onlynine of the 28 patients with CHD and in none ofthe nine patients with miscellaneous forms of heartdisease. This latter group had the largest number of

patients with congestive heart failure. It may wellbe that reduction in cardiac work load during sleepis a contributory factor to the salutary effect onventricular arrhythmia. During sleep, in addition tothe slowing in rate, there is a reduction in bloodpressure. In the normal young adult blood pressurereaches a nadir within 1.5 to 2.5 hr after onset ofsleep; thereafter, it rises through the night to attainits basal level shortly before awaking.21 In normalas well as hypertensive patients, the fall in bloodpressure has been ascribed to reduction in totalperipheral resistance.22 A modest decrease incardiac output has also been noted.23 The abovecited hemodynamic changes do not coincide tem-porally with the abatement of arrhythmia. Thedecrease in VPBs is noted with the very onset ofsleep at a time when measurable hemodynamicalterations have not yet occurred. If a change inhemodynamic load was of importance in thepresent study, one would have anticipated phasicvariation in ectopic activity throughout sleepcoinciding with REM state, during which cardiacwork is augmented. This was not observed.Hemodynamic variables are probably not thedecisive factors accounting for the lessening inVPBs during sleep.Heart rate changes and other cardiac alterations

during sleep are mediated by both sympathetic andparasympathetic neural outflow24' 25 and are abol-ished after complete cardiac denervation.2' Cardiacautomaticity and excitability, like heart rate, arecarefully tuned by sympathetic discharge.26 Becauseof partial sympathetic withdrawal during sleepthere is a rise in threshold to artificial pacing. In apatient with complete heart block, pacing ceasedwith sleep only to be restored immediately uponawaking.27 In the present study, the observedreduction during sleep in VPB incidence and grademay therefore be related to lessened sympathetictone. If this is indeed a key factor, it is hard toaccount for the failure of large doses of propranololto reduce VPBs in three patients, though such aresult was observed with sleep alone.While the specific neural mechanism responsible

for altered ventricular ectopic activity with sleepremains to be defined, there can be little doubt thathigher nervous activity is implicated. Furthermore,it appears that this neural effect is highly potent, forin a number of patients sleep reduced VPBs whenantiarrhythmic drugs such as quinidine, procaineamide, and propranolol were without effect. It maybe that for many patients with paroxysmal ventricu-lar arrhythmia the pharmacologic focus should be


700


Dow

nloaded from


SLEEP AND VPBs

in restraining the neurophysiologic trigger ratherthan in attempting to protect the cardiac target.

References1. MACWILLIAM JA: Blood pressure and heart action in

sleep and dreams: Their relation to hemorrhages,angina and sudden death. Br Med J 2: 1196, 1923

2. LOWN B, WOLF M: Approaches to sudden death fromcoronary heart disease. Circulation 44: 130, 1971

3. DUCKSWORTH D: On insomnia and other troublesconnected with sleep in persons of gouty disposition.Brain 4:145, 1881

4. ASSERINSKY E, KLEITMAN N: Regularly occurringperiods of eye motility and concurrent phenomenaduring sleep. Science 118: 273, 1953

5. DEMENT W, KLEITMAN N: Cyclic variations in EECduring sleep and their relation to eye movements,body motility and dreaming. Electroencephalogr ClinNeurophysiol 9: 673, 1957

6. NOWLIN JB, TROYER WG JR, COLLINS WS, SILVERMANG, NICHOLAS CR, MCINTOSH HD, EsTEs EH JR,BOGDONOFF MD: The association of nocturnal anginapectoris with dreaming. Ann Intern Med 63: 1040,1965

7. KARACAN I, WILLIAMS RL, TAYLOR WJ: Sleepcharacteristics of patients with angina pectoris.Psychosomatics 10: 280, 1969

8. ROSENBLATr G, ZWILLING G, HARTMANN E: Electro-cardiographic changes during sleep in patients withcardiac abnormalities. (abstr) Psychophysiology 6:233, 1969

9. SMITH R, JOHNSON L, ROTHFELD D, ZIR L, THARP B:Sleep and cardiac arrhythmias. Arch Intern Med130: 751, 1972

10. SHAHAWAY EM: Arrhythmias and varieties of sleep.(Letter) N Engl J Med 282: 815, 1970

11. KLErrMAN N: Sleep and wakefulness. Chicago,University of Chicago Press, 1963

12. HAN J, DETRAGLIA J, MILLET D, MOE GK: Incidenceof ectopic beats as a function of basic rate in theventricle. Am Heart J 72: 632, 1966

13. HAN J, MILLET D, CHIZZONITI B, MOE GK: Temporaldispersion of recovery of excitability in atrium and

ventricle as a function of heart rate. Am Heart J 71:481, 1966

14. LOWN B, LEVINE SA: The carotid sinus: Clinical valueof its stimulation. Circulation 23: 766, 1961

15. ROTHBERGER CJ, WINTERBERG UH: Uber scheinbareVaguslihmung bei Muskarin, Physostigmin undanderen Ciften sowie bei interkardialer Drucksteiger-ung. Pfluegers Arch ges Physiol 132: 233, 1910

16. VASSALLE M, CREENSPAN K, HOFFMAN BF: An analysisof arrhythmias induced by ouabain in intact dogs.Circ Res 13: 132, 1963

17. LOWN B, EHRLICH L, LIPSCHULTZ B, BLAKE J: Effect ofdigitalis in patients receiving reserpine. Circulation24: 1185, 1961

18. WITTENBERG SM, LOWN B: Cardioversion and digita-lis: IV. Effect of beta adrenergic blockade. Circula-tion 39: 29, 1969

19. PiTT B, KURLAND CS: Use of edrophonium chloride(Tensilon) to detect early digitalis toxicity. Am JCardiol 18: 557, 1966

20. LANGENDORF R, PICK A, WINTERNITZ M: Mechanism ofintermittent ventricular bigeminy: I. Appearance ofectopic beats dependent upon length of ventricularcycle, the "rule of bigeminy." Circulation 11: 422,1955

21. SNYDER F, HOBSON JA, MOFRISON DF, GOLDFRANK F:Changes in respiration, heart rate and systolic bloodpressure in human sleep. J Appl Physiol 19: 417,1964

22. BRisTow JD, HONOUR AJ, PICKERING TG, SLEIGHT P:Cardiovascular and respiratory changes during sleepin normal and hypertensive subjects. Cardiovasc Res3: 476, 1969

23. KHATRI IM, FREIS ED: Hemodynamic changes duringsleep. J Appl Physiol 22: 867, 1967

24. BAUST W, BOHNERT B: The regulation of heart rateduring sleep. Exp Brain Res 7: 169, 1969

25. SAMAAN A: La frequence cardiaque du chien endifferentes conditions experimentales d'activite et derepos. C R Soc Biol (Paris) 115: 1383, 1934

26. HOFFMAN BF, CRANEFIELD PF: Electrophysiology ofthe Heart. New York, McGraw-Hill Book Co, Inc,1960

27. SOMERDIKE JM, OSTEERMILLER WE, CARATA JM,SALYER JM: (Letter) JAMA 215: 980, 1971

Circulation, Volsme XLVIII, October 1973

701


Dow

nloaded from


BERNARD LOWN, MARK TYKOCINSKI, ANDRE GARFEIN and PHILLIP BROOKSSleep and Ventricular Premature Beats

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1973 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.48.4.691

1973;48:691-701Circulation.

http://circ.ahajournals.org/content/48/4/691Wide Web at:

The online version of this article, along with updated information and services, is located on the World

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer in the

Permissions in the middle column of the Web page under Services. Further information about this process is availableOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Circulation Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:


Dow

nloaded from

http://circ.ahajournals.org/content/48/4/691

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


Sleep and Ventricular Premature -...

Documents

Transcript of Sleep and Ventricular Premature -...